Last updated June 2021

Personal details

A few things about me: I love reading books about behavioral science, art and literature. Check my goodreads for books I have read. Sometimes I wrote review about books as well. I never get bored of playing board-game with friends. Lastly, I love learning. Recently I push myself to write more what I have learnt so far. Please check my medium to read my summaries of what I have learnt.

I have more than 4 years of experience in Data Analytics area ranging from extracting data, processing data to build reports and models. My main focus domain is digital marketing science & customer insight. Also I am active contributor in AI for good (Omdena) platform taking advantage of Data Science in tackling social issues.

Please find my resume here & more detailed portfolio below:

  1. Proposal: Ranking influencers engagement, Scoring Fans' comment, Detect trending topics (Creatory: Aug 2020 — present)

The projects' goals are to propose then build data-driven features for upcoming social network platform which connect Fans and Influencers.

  • Key responsibilities:

Propose data-driven product feature

Research & develop algorithm for each feature

Work with Engineer team to decide which features need to collect data.

  • Results: Prototype of above features

2. Renewable energy in Nigeria (Omdena: Nov 2019 — Jan 2020)

In this project, Renewal Africa 365 — a non-profit organization partners with Omdena to predict the most conducive places for Solar Panel Installations in Nigeria.

  • Key responsibilities:

Lead sub-task of working on estimating energy demand in Nigeria

Extracted & QA satellite image data about night light & population in Nigeria from Google Earth Engine via Python.

Cluster energy demand area using DBscan.

Ranked priority of clusters

  • Results:

Visualization of clusters of energy demand in Nigeria as well as cluster point in csv format as the input for another sub-task groups.

Deliver ranking cluster point in csv format to partner

3. Deployed time series forecast for web traffic in Google Cloud (Hello Health Group: Feb — Apr 2020)

The project is to forecast daily organic traffic of the company’ site then predict the Google’s algorithm update time then Marketing & Content team to update article corresponding.

  • Key responsibilities:

Extract organic traffic data from Google Big Query

Forecast daily organic traffic using Prophet

Anomaly detection of organic traffic using Luminol

Work with Data Engineer to ETL data.

Schedule daily alert (if any) about upcoming Google’ algorithm change using Google Cloud Function.

  • Results:

Daily alert via email to stakeholders about upcoming Google’ algorithm change (if any).

4. Readability score (Hello Health Group: Apr — Jun 2020)

Readability score is a number that tells how easy it will be for someone to read a particular piece of text. The project is to research then build readibility score for multiple languages.

  • Key responsibilities:

Research & develop readability score for articles in English, Bahasa, Vietnamese, Hindi language.

Automate scoring in Google Cloud when new articles coming in.

Work with Business Intelligence to build dashboard in Google Data Studio.

  • Results:

Result of article’s readability score updating frequently in Google Big Query

Dashboard of readability score in Google Data Studio.

5. Customer lifetime value (Hello Health Group: Jun — Jul 2020)

The lifetime value of an user represents the total amount of money an user is expected to bring to the company via appearing ads on company’s article. The project goal is to estimate the lifetime value of user for the next 6 months.

  • Key responsibilities:

Extract data about Google ads in Google Bigquery

Predicting customer lifetime value using probabilistic model

  • Results:

Table of customer lifetime value in the next 6 months in Google Bigquery.

6. Build ESG score & correlation score (Sentifi: Jan 2019 — Jan 2020)

The projects are to build formula for ESG (Environmental Societal and Governance score) basing on social listening data & Correlation score basing on stock price and social listening data.

  • Key responsibilities:

Extract data then build scores

Defend formula with C-level

QA scores result from API built by Backend team.

  • Results:

Deliver formula to Backend engineers.

Document scores for external customer.

7. Build internal python package to extract internal data (Sentifi: Nov 2019 — Jan 2020)

The project is to build python library for internal member in Data Operation team to extract raw and aggregated data in multiple sources.

  • Key responsibilities:

Write python module to extract data from Elasticsearch database

  • Results:

Python package for internal user in Data operation team.

8. Customer data platform (Tiki Jan — Nov 2018)

The project is to build platform for internal stakeholder finding insights and easily interacting with customer data.

  • Key responsibilities:

Define & ETL customer data from multiple sources in Google Cloud

Define features of Platform

Train stakeholders about how to use platform.

  • Results:

Friendly user interface platform for Marketing and Category teams.

Feral cat trying to make sense people and the world around